Low-Flow Estimation and Prediction

Full text

(1)Manual on Low-flow Estimation and Prediction WMO-No. 1029. Operational Hydrology report No. 50.

(2) Manual on Low-flow Estimation and Prediction Operational Hydrology Report No. 50. WMO-No. 1029.

(3) German IHP/HWRP National Committee. BfG – Federal Institute of Hydrology. Layout: IHP/HWRP Secretariat Federal Institute of Hydrology Am Mainzer Tor 1 56068 Koblenz • Germany Tel.: +49 (0)261/1306-5435 Fax: +49 (0)261/1306-5422 http://ihp.bafg.de German National Committee for the International Hydrological Programme (IHP) of UNESCO and the Hydrology and Water Resources Programme (HWRP) of WMO Koblenz 2009 Scientific Editors: Alan Gustard and Siegfried Demuth Linguistic Editor: Susan Parker. WMO-No. 1029 © World Meteorological Organization, 2008 The right of publication in print, electronic and any other form and in any language is reserved by WMO. Short extracts from WMO publications may be reproduced without authorization, provided that the complete source is clearly indicated. Editorial correspondence and requests to publish, reproduce or translate this publication in part or in whole should be addressed to: Chairperson, Publications Board World Meteorological Organization (WMO) 7 bis, avenue de la Paix P.O. Box No. 2300 CH-1211 Geneva 2, Switzerland Tel.: +41 (0) 22 730 84 03 Fax: +41 (0) 22 730 80 40 E-mail: [email protected] ISBN 978-92-63-11029-9 Cover photograph: M. Hall. NOTE: The designations employed in WMO publications and the presentation of material in this publication do not imply the expression of any opinion whatsoever on the part of the Secretariat of WMO concerning the legal status of any country, territory, city or area, or of its authorities, or concerning the delimitation of its frontiers or boundaries. Opinions expressed in WMO publications are those of the authors and do not necessarily reflect those of WMO. The mention of specific companies or products does not imply that they are endorsed or recommended by WMO in preference to others of a similar nature which are not mentioned or advertised..

(4) Contents. III. Contents Foreword Preface Acknowledgement List of Authors Summary . V VII VIII IX X. 1. Introduction 1.1 Objectives 1.2 Background 1.3 Manual structure 1.4 Methodologies not included in the Manual 1.5 Guidelines for national practices. 13 13 13 14 15 15. 2. Estimating, Predicting and Forecasting Low Flows 2.1 Introduction 2.2 Low-flow information 2.3 Design issues 2.4 Previous studies 2.5 Dissemination of results 2.6 Key principles in low-flow design. 16 16 16 17 20 20 20. 3. Hydrological Data 3.1 Introduction 3.2 Data for low-flow analysis 3.3 Preparation of good-quality data for low-flow analysis . 22 22 22 26. 4. Processes and Regimes 4.1 Introduction 4.2 Processes causing low flows 4.3 Low flows in different hydrological regimes. 36 36 36 39. 5. Low-Flow indices 5.1 Introduction 5.2 Base-flow index 5.3 Recession analysis 5.4 Low-flow statistics 5.5 Operational applications. 43 43 43 44 47 48. 6. The Flow-Duration curve 6.1 Definition and derivation 6.2 Standardization 6.3 Durations and seasons 6.4 Percentiles as low-flow indices 6.5 Applications of the FDC. 50 50 51 52 53 53. 7. Extreme value analysis 7.1 Introduction 7.2 Example data for at-site low-flow frequency analysis 7.3 Introduction to frequency analysis 7.4 Extreme value selection 7.5 Distribution functions 7.6 Parameter estimation methods 7.7 Estimation of the T-year event 7.8 Application of the Weibull distribution for low-flow frequency analysis 7.9 Regional frequency analysis. 57 57 57 58 60 61 63 64 65 67.

(5) IV. Contents. 8. Streamflow deficit 8.1 Introduction 8.2 Definition and derivation 8.3 Application: Determination of streamflow deficit characteristics 8.4 Definitions of low-flow and deficit characteristics: National standards in Germany . 71 71 71 73 74. 9. Estimating low flows at ungauged sites 9.1 Introduction 9.2 Empirical methods 9.3 Statistical methods – regionalization 9.4 Catchment modelling 9.5 Use of local data. 77 77 77 78 81 84. 10. Estimating low flows in artificially influenced rivers 10.1 Introduction 10.2 Inventory of artificial influences 10.3 Artificial influences 10.4 Estimating artificially influenced low flows 10.5 Flow naturalization 10.6 Climate and land-use change. 88 88 88 89 92 94 95. 11. Low-flow forecasting 11.1 Introduction 11.2 Short-term forecasting 11.3 Medium-term forecasting 11.4 Purpose of long-term forecasting 11.5 Basic modelling techniques for forecasting 11.6 Conclusions. 98 98 99 101 102 104 107. 12. Case studies 12.1 Summary of the case studies 12.2 Transboundary rivers 12.3 Catchment-based water resources decision-support tool in the United Kingdom 12.4 Low-flow management issues in the United Kingdom 12.5 Real-time management of environmental flow requirements for the Thukela River in South Africa 12.6 Regionalized resources models for small-scale hydropower: India and Nepal 12.7 Residual flow estimation and hydropower: Norway. 108 108 109 115 117 120 123 125. 13. Recommendations and conclusions 13.1 Introduction 13.3 Operational applications 13.4 Capacity-building . 129 129 129 131. Abbreviations Operational Hydrology Reports. 134 136.

(6) FOREWORD. V. Foreword. The Commission for Hydrology (CHy) decided at its twelfth session (Geneva, October 2004) to prepare a manual on low-flow estimation and prediction to meet the identified needs of National Hydrological Services. The Manual on Low-flow Estimation and Prediction, which consists of 13 chapters, was drafted by the Open Panel of CHy Experts (OPACHE) on Disaster Mitigation – Floods and Droughts (Hydrological Aspects). The list of authors is provided on page IX. The drafting team was initially led by Mr Siegfried Demuth (Germany), then by Mr Bruce Stewart (Australia), who also took part in the review process. Mr Charles Pearson (New Zealand) and Mr Syed Moin (Canada) were the reviewers.. The Manual’s objective is to publish state-of-the-art analytical procedures for estimating and predicting low river flows at all sites, regardless of the availability of observational data. The Manual will be useful for many applications, including water resources planning, effluent dilution estimates and water resources management during low-flow conditions. It is a great pleasure to express WMO’s gratitude to the authors, the reviewers and to the president of CHy, Mr Bruce Stewart, for their efforts during the preparation of the Manual.. (M Jarraud) Secretary-General.

(7)

(8) Preface. VII. Preface. More often than not, the thought of rivers conjures up the image of high water levels and even floods. However, many communities depend on the availability of water through non-regulated river systems for their water supply. Therefore, periods of low flow are critical for managing their water resources. Similarly, it is the function and role of dams to level out the fluctuations of high and low flows and to provide a balanced water supply to meet demands. Knowledge of low-flow periods is therefore fundamental for reservoir design and determining allocations. Many factors have an impact on the low-flow regimes of rivers. Whenever we change our land uses, we change the way in which water interacts with the landscape, and this can affect the water available in rivers, lakes and dams. Most nations are experiencing population growth, resource depletion and the overextraction of water. Low flows are critical elements in terms of meeting demands for often competing uses and requirements.. More recently, there have been growing concerns over the relationship between water and the environment, and many counties are conducting research into the requirements for, and delivery of, environmental flows. The topic of this Manual, namely low-flow estimation, is thus of paramount importance for the development and implementation of sustainable water resources management practices. I highly recommend this Manual and congratulate its contributors on the guidance provided therein. I note that the Manual does not have the full geographical coverage that the Commission for Hydrology would like to see in its publications. However, we consider this a challenge for future publications and will seek your support in addressing the issue when we next address the topic. We would, of course, be interested in any feedback on the use and implementation of this Manual.. (Bruce Stewart) President of the Commission for Hydrology.

(9) VIII. acknowledgement. Acknowledgement. Alan Gustard (UK) and Siegfried Demuth (Federal Republic of Germany) acted as editors-in-chief for the manual. The reviewers were Mr. Charles Pearson (New Zealand) and Mr. Syed Moin (Canada). The activities were carried out through the Open Panel of CHy Experts (OPACHE) on Disaster Mitigation – Floods and Droughts.. The authors wish to thank the following entities and individuals for their cooperation and the information kindly provided. Chapter 2: Duncan Reed of DWRconsult (UK) for helpful advice on strategic aspects of hydrological design. Chapter 4: The Water Survey of Canada for streamflow data from the HYDAT database available at: http://www.wsc.ec. gc.ca/products/hydat/main_e.cfm?cname=hydat_e.cfm. Chapter 5: Gunnar Wollan, senior engineer, and Maria Staudinger, a visiting student, at the Department of Geosciences, University of Oslo, for their contribution to the calculation of recession parameters in section 5.3.. Chapter 6: The Norwegian Water Resources and Energy Directorate, the Centre for Ecology and Hydrology (United Kingdom) and the US Geological Survey for making the time series of river flows available for determining flow-duration curves; and the Devon Area Regulatory and Technical (Water Resources) Team of the Environment Agency (England and Wales) for providing the example of an abstraction licence for a fish farm. Chapter 8: The Norwegian Water Resources and Energy Directorate and the US Geological Survey for the time series used in the derivation of streamflow deficit, and Walter Finke of the Federal Institute of Hydrology for his example of German standards for determining low-flow and deficit characteristics. Chapter 12: John Metzger from the Mekong River Commission for providing reports on the Mekong; Turid-Anne Drageset and Astrid Voksø at the Norwegian Water Resources and Energy Directorate for their contribution to the residual flow case study; and the Alternate Hydro Energy Centre of the Indian Institute of Technology, Roorkee, the Himachal Pradesh Energy Development Agency (HIMURJA), India, and the Centre for Ecology and Hydrology for their contribution to the hydropower case study..

(10) list of authors. IX. List of Authors. Chapter 1: Introduction Alan Gustard and Siegfried Demuth. Chapter 8: Streamflow deficit Hege Hisdal. Chapter 2: Estimating, predicting and forecasting low flows Alan Gustard and Siegfried Demuth. Chapter 9: Estimating low flows at ungauged sites Andy Young, Denis Hughes and Siegfried Demuth. Chapter 3: Hydrological data Gwyn Rees Chapter 4: Processes and regimes Kerstin Stahl, Henny A.J. van Lanen and Stefan Uhlenbrook Chapter 5: Low-flow indices Alan Gustard and Lena Tallaksen Chapter 6: The flow-duration curve Hege Hisdal and Alan Gustard Chapter 7: Extreme value analysis Lena Tallaksen and Guna A. Hewa. Chapter 10: Estimating low flows in artificially influenced rivers Alan Gustard and Andy Young Chapter 11: Low-flow forecasting Trevor Daniell, Rory Nathan, Francis Chiew and Alexander Osti Chapter 12: Case studies Andy Young, Kerstin Stahl, Gwyn Rees, Alan Gustard, James Dent, Hege Hisdal and Denis Hughes Chapter 13: Recommendations and conclusions Alan Gustard, Siegfried Demuth, Roland Price and Stefan Uhlenbrook.

(11) X. Summary. Summary The Manual’s objective is to publish state-of-the-art analytical procedures for estimating and predicting low river flows at all sites, regardless of the availability of observational data. The Manual will be used for estimating and predicting low flows for many applications, including water resources planning, effluent dilution estimates and water resources management during low-flow conditions. This Manual will be one of the technical guidance documents included in the WMO Quality Management Framework in Hydrology and discussed by the thirteenth session of the Commission for Hydrology. Chapters 1 and 2 of the Manual introduce the objectives, structure and major issues involved in predicting and forecasting low flows. Chapter 3 discusses the data requirements for low-flow estimation, including river flow and associated basin properties, for example, soil type, hydrogeology and climate. Chapter 4 presents key low-flow processes and the resulting wide range of low-flow response, an understanding of which is essential for analysing and interpreting low-flow information. Chapter 5 describes simple low-flow indices, including the 95 percentile exceedance discharge and base-flow and recession characteristics. Chapters 6, 7 and 8 provide. step-by-step guidelines on estimating the flow-duration curve, extreme value distributions and the analysis of streamflow deficits, respectively. Chapter 9 describes a range of methods for estimating low flows at ungauged sites, including the use of short and nearby flow records to reduce the uncertainty of flow estimation. Chapter 10 presents key practical problems of how to estimate low flows in rivers influenced by artificial controls, such as abstractions, effluent returns and impoundments. Chapter 11 describes the main applications for which forecasts of low flows are required and presents methodologies for forecasting flows on a range of timescales. Chapter 12 presents a number of case studies on, among others, transboundary issues, a water resources decision-support tool, a regional approach to estimating small-scale hydropower and the estimation of residual flows below abstraction points. Chapter 13 presents some significant conclusions and recommendations relating to data collection and capacity-building. Together with the techniques presented in the rest of the Manual, it is hoped that these conclusions and recommendations will reduce the uncertainty of estimating low flows and improve methods, for the benefit of all users.. Résumé Le présent manuel a pour objet de faire largement connaître les méthodes analytiques les plus avancées en matière d’estimation et de prévision des débits d’étiage en tous points des cours d’eau, quelles que soient les données d’observation disponibles, et cela dans la perspective de nombreuses applications, dont la planification des ressources en eau, les estimations de la dilution des effluents et la gestion des ressources en eau pendant les périodes d’étiage.. tivement des indications, étape par étape, sur l’estimation de la courbe des débits classés, la distribution des valeurs extrêmes et l’analyse des déficits d’écoulement fluvial.. Les chapitres 1 et 2 portent sur les objectifs, la structure et les principales questions relatives à la prédétermination et à la prévision des débits d’étiage. Le chapitre 3 est consacré aux données nécessaires pour l’estimation de ces débits, y compris l’écoulement fluvial et les propriétés des bassins connexes (types de sols, hydrogéologie, climat, etc.).. Au chapitre 9 sont décrites un certain nombre de méthodes d’estimation du débit d’étiage sur des sites non jaugés, y compris l’utilisation de relevés de débit de courte durée et exécutés à proximité afin de réduire l’incertitude des valeurs estimées. Au chapitre 10 sont présentés les principaux problèmes pratiques que pose l’estimation des débits d’étiage dans les cours d’eau soumis à l’influence de contrôles artificiels (dérivations, renvois d’effluents, retenues, etc.). Au chapitre 11 sont abordées les principales applications nécessitant des prévisions du débit d’étiage et les méthodes de prévision du débit à diverses échelles de temps. Au chapitre 12 sont présentées un certain nombre d’études de cas portant notamment sur des questions de caractère transfrontalier, un outil d’aide à la décision pour les ressources en eau, une approche régionale de l’estimation de l’énergie hydroélectrique à petite échelle et l’estimation des débits résiduels en aval des points de dérivation.. Au chapitre 4 sont présentés les principaux processus liés aux étiages ainsi que le large éventail des réponses possibles, qu’il est indispensable de bien comprendre pour analyser et interpréter les informations relatives aux étiages. Au chapitre 5 sont indiqués les indices d’étiage simples, y compris le débit de dépassement correspondant au quatre-vingt-quinzième percentile et les caractéristiques propres aux débits de base et aux décrues. Les chapitres 6, 7 et 8 fournissent respec-. Au chapitre 13 figurent quelques conclusions et recommandations importantes concernant la collecte des données et le renforcement des capacités. Il est à espérer qu’à l’instar des techniques présentées dans le reste du manuel, ces conclusions et recommandations contribueront à atténuer l’incertitude propre aux valeurs estimées des débits d’étiage et à améliorer les méthodes employées, pour le plus grand profit de tous les utilisateurs.. Ce manuel sera l’un des documents techniques d’orientation qui seront pris en compte dans le Cadre de référence de l’OMM pour la gestion de la qualité en hydrologie et qui seront examinés par la Commission d’hydrologie à sa treizième session..

(12) Summary. XI. Резюме Цель настоящего Наставления заключается в том, чтобы опубликовать современные аналитические процедуры для расчета и прогнозирования низкого речного стока на всех объектах, независимо от наличия данных наблюдений. Наставление будет использоваться для расчета и прогнозирования низкого стока для многих применений, включая планирование водных ресурсов, оценку разбавления стоков и управление водными ресурсами в условиях низкого стока. Это Наставление будет являться одним из технических руководящих документов, включенных в Структуру управления качеством ВМО в области гидрологии, которые были обсуждены на тринадцатой сессии Комиссии по гидрологии. В главах 1 и 2 Наставления представлены цели, структура и основные вопросы, связанные с предсказанием и прогнозированием низкого стока. В главе 3 рассматриваются требования к данным для оценки низкого стока, включая речной сток и связанные с ним свойства бассейнов, например, тип почвы, гидрогеология и климат. В главе 4 представлены ключевые процессы низкого стока и получаемый широкий диапазон реагирования низкого стока, понимание которого необходимо для анализа и толкования информации о низком стоке. В главе 5 описываются простые индексы низкого стока, включая 95-процентильное превышение расхода и характеристики базисного стока и истощения стока. В главах 6, 7 и 8 представлены поэтапные руководящие. принципы для расчета кривой продолжительности стока, распределения экстремальных значений и анализа дефицита речного стока соответственно. В главе 9 описывается ряд методов для оценки низкого стока на участках, где не проводятся постоянные измерения, включая использование краткосрочных данных о стоке с близлежащих станций в целях снижения неопределенности при расчете стока. В главе 10 описываются ключевые практические проблемы, касающиеся расчета низкого стока в реках под влиянием искусственных контрольных сечений, таких как отводы, возврат истоков и запруживание. В главе 11 описаны основные применения, для которых требуются прогнозы низкого стока, и представлены методики для прогнозирования стока по диапазону временных масштабов. В главе 12 представлен ряд тематических исследований, посвященных, среди прочего, трансграничным вопросам, инструменту, поддерживающему принятие решений, касающихся водных ресурсов, региональному подходу к оценке гидроэнергии в небольших масштабах и расчетов остаточного стока ниже точек отвода. В главе 13 представлены некоторые важные выводы и рекомендации, касающиеся сбора данных и наращивания потенциала. Наряду с методами, представленными в остальной части Наставления, выражается надежда, что эти выводы и рекомендации позволят понизить неопределенность при расчете низкого стока и улучшить методику на благо всех пользователей.. Resumen El objetivo del Manual es publicar los procedimientos analíticos más avanzados para estimar y predecir el caudal de estiaje en cualquier lugar, con independencia de la disponibilidad de datos de observación. El Manual se usará para estimar y predecir el caudal de estiaje, lo que facilitará la realización de numerosas aplicaciones -tales como la planificación de los recursos hídricos, la estimación de la dilución de efluentes y la gestión de los recursos hídricos- cuando se dé un caso de estiaje. Este Manual será uno de los documentos de orientación técnica que se incluyan en el Marco de gestión de la calidad de la OMM – Hidrología y que examine la Comisión de Hidrología en su decimotercera reunión. En los capítulos 1 y 2 del Manual se enuncian los objetivos, la estructura y las principales cuestiones pertinentes de la predicción y previsión del caudal de estiaje. En el capítulo 3 se estudian los datos que se necesitan para realizar estimaciones del caudal de estiaje, entre ellos el caudal fluvial y las propiedades relacionadas de las cuencas hidrográficas, como por ejemplo el tipo de suelo, la hidrogeología y el clima. El capítulo 4 presenta procesos fundamentales del caudal de estiaje y la amplia gama resultante de respuestas existente en ese ámbito, cuya comprensión es esencial para analizar e interpretar la información sobre el caudal de estiaje. En el capítulo 5 se describen índices sencillos de caudal de estiaje, en particular el percentil 95 de excedencia de caudales y las características del caudal de base y de la recesión. En los capítulos 6, 7 y 8 figuran directrices de fases progresivas sobre la estimación. de la curva de duración de caudales, la distribución de los valores extremos y el análisis de los déficits de flujo fluvial respectivamente. En el capítulo 9 se describe un conjunto de métodos para estimar el caudal de estiaje en lugares que no han sido aforados, entre los que cabe citar la utilización de registros de caudales de períodos cortos y de lugares cercanos para reducir la incertidumbre de su estimación. En el capítulo 10 se presentan problemas prácticos fundamentales de la estimación del estiaje en ríos que están bajo la influencia de controles artificiales, tales como las extracciones, los retornos de efluentes o las retenciones. En el capítulo 11 se describen las principales aplicaciones para las que se necesitan las predicciones del caudal de estiaje y se presentan métodos para predecir el caudal en diversas escalas temporales. En el capítulo 12 se presentan varios estudios de casos sobre, entre otras cosas, cuestiones transfronterizas, así como un instrumento de apoyo para la adopción de decisiones sobre los recursos hídricos, un enfoque regional para estimar la energía hidroeléctrica en pequeña escala y la estimación de los caudales residuales por debajo de los puntos de extracción. En el capítulo 13 se presentan algunas conclusiones y recomendaciones importantes relacionadas con la recogida de datos y la creación de capacidad. Se espera que estas conclusiones y recomendaciones, junto con las técnicas presentadas en el resto del Manual, sirvan para reducir la incertidumbre de las estimaciones del caudal de estiaje y para mejorar los métodos utilizados a tal fin, en beneficio de todos los usuarios..

(13)

(14) INTRODUCTION. 13. 1. Introduction. 1.1 . Objectives. One of the aims of the World Meteorological Organization (WMO) is to promote the standards of meteorological and hydrological observations and ensure the consistent and appropriate analysis of environmental data. A number of WMO guidance manuals have been published which describe in detail the practices, procedures and specifications that WMO Members are invited to implement. This Manual is a contribution to this task, and its key objective is to publish state-of-the-art analytical procedures for estimating, predicting and forecasting low river flows at all sites, regardless of the availability of observational data. The Manual will be used by operational agencies to predict and forecast low flows for a wide range of applications, including national and regional water resources planning, abstraction management, public water-supply design, instream flow determination, effluent dilution estimates, navigation, the run design of river hydropower schemes, the design of irrigation schemes and water resources management during low-flow conditions. The Manual will also make it possible to mitigate the hydrological impacts of low flows and facilitate the design of flow-monitoring networks. The Manual includes a description of current methodologies used for low-flow prediction and forecasting and practical examples of operational applications. It addresses the estimation of low flows from continuous flow records, at ungauged sites and from short flow records including natural and artificially influenced catchments. The Manual will be of value primarily to hydrologists from operational agencies where low-flow estimation procedures are being developed or updated and to professionals teaching short technical courses and first degree and Master of Science courses in applied hydrology. 1.2 . Background. The Manual is part of a new series of WMO publications consisting of several manuals and guidelines. within the WMO Quality Management Framework in Hydrology. The manuals all have a practical approach to hydrological and water resources design and are targeted at meeting the needs of National Hydrological and Meteorological Services. These manuals include theoretical information only when it is needed for a better understanding of the subject. In this series, manuals are planned on the following subjects: flood forecasting; probable maximum precipitation/probable maximum floods; design floods; low-flow estimation; and water resources assessment. These manuals will complement the information available in the Guide to Hydrological Practices (WMO-No. 168) and the WMO Technical Regulations (WMO-No. 49). This Manual complements the Hydrological Operational Multipurpose System (HOMS) established by WMO for the transfer of technology in hydrology and water resources. This technology is usually in the form of descriptions of hydrological instruments, technical manuals or computer programs. The material included in HOMS is used operationally by the Hydrological Services of WMO Members. This ensures that the technology transferred is not only ready for use, but also works reliably. Participating countries designate a HOMS National Reference Centre (HNRC), usually in the National Hydrological Service. This centre provides national components for use in HOMS, handles national requests for the HOMS components to be supplied by other HNRCs, advises users on HOMS, and generally coordinates and publicizes HOMS activities. The terminology used in this report is consistent with the UNESCO and WMO International Glossary of Hydrology (UNESCO/WMO, 1992) and Hydrological Drought – Processes and Estimation Methods for Streamflow and Groundwater (Tallaksen and van Lanen, 2004). The latter also provides a broader perspective on a wide range of drought issues, including the hydrological processes controlling drought response, and methods for analysing river flows and groundwater droughts, stream ecology and flow management..

(15) 14. 1.3 . INTRODUCTION. Manual structure. The Manual presents a range of different techniques for analysing hydrological data to provide operational information for low-flow prediction and forecasting. These are summarized in Table 1.1, which lists different ways of describing the low-flow regime of a river, the specific property of the analysis technique, data requirements and some common applications. The Manual provides step-by-step guidance for calculating these low-flow properties from recorded flow data and presents methods for estimating them at ungauged sites and for forecasting low flows on a range of time scales. Analysis techniques are illustrated by practical examples of their use in operational hydrology. This chapter and Chapter 2 of the Manual introduce the objectives, structure and key issues involved in predicting and forecasting low flows. These are followed by Chapter 3, which provides a comprehensive review of the data requirements for low-flow estimation, including river-flow and associated catchment properties, for example, soil type, hydrogeology and climate. Chapter 4 presents key low-flow processes and the resulting wide range of low-flow response, an understanding of which. is essential for analysing and interpreting low-flow information. Chapter 5 describes simple low-flow indices, including the 95 percentile exceedance discharge and base-flow and recession characteristics. Chapters 6, 7 and 8 provide step-by-step guidelines for estimating the flow-duration curve, extreme value distributions and the analysis of streamflow deficits, respectively. Chapter 9 describes a range of methods for estimating low flows at ungauged sites, including the use of short and nearby flow records to reduce the uncertainty of flow estimation. The key practical problems of how to estimate low flows in rivers influenced by artificial controls, such as abstractions, effluent returns and impoundments, are presented in Chapter 10. Chapter 11 describes the main applications for which low-flow forecasts are required and presents methodologies for forecasting flows on a range of timescales. Chapter 12 presents a number of case studies, which include the topics of transboundary rivers, a water resources decisionsupport tool, a regional approach to estimating smallscale hydropower and the estimation of residual flows below abstraction points. It is hoped that some of the procedures described in the analytical chapters will thereby be placed in a broader context.. Table 1.1 Summary of low-flow regime measures (based on Gustard and others, 1992) Regime measured. Property described. Data employed. Applications. Mean flow. Arithmetic mean of the flow series. Daily or monthly flows. Resource estimation. Coefficient of variation in annual mean flow. Standard deviation of annual mean flow divided by mean flow. Annual mean flow. Understanding of regime interannual variability; definition of carryover storage requirements. Flow-duration curve. Proportion of time a given flow is exceeded. Daily flows or flows averaged over several days, weeks or months. General regime definition; licensing abstractions (water rights) or effluents (discharge consents); hydropower design. Annual minimum series. Annual lowest flows (of a given duration). Annual minimum flows – daily or averaged over several days. Drought return period; preliminary design of major schemes; first step in some storage/yield analyses. Streamflow deficit durations. Frequency with which the flow remains below a threshold for a given duration. Periods of low flows extracted from the hydrograph followed by a statistical analysis of durations. More complex water quality problems, such as fisheries and amenity, navigation; general indication of drought frequency. Streamflow deficit volumes. Frequency of requirement for a given volume of “make-up” water to maintain a threshold flow. Same as above, except the analysis focuses on the volume below the threshold. Preliminary design of regulating reservoirs; general indication of drought frequency. Recession indices. Rate of decay of hydrograph. Daily flows during dry periods. Short-term forecasting; hydrogeological studies; modelling. Base-flow index. Proportion of total flow which comes from stored catchment sources. Daily flows. Hydrogeological studies; preliminary recharge estimation.

(16) INTRODUCTION. Chapter 13 presents some key conclusions and recommendations relating to data collection, operational hydrology and capacity-building. Together with the techniques presented in this Manual, it is hoped that these will lead to a reduction in the uncertainty of estimating low flows and to improved methods for transferring this information to a range of stakeholders. 1.4 . Methodologies not included in the Manual. Although the most common procedure for mitigating the impact of low flows is to provide reservoir storage, storage yield relationships for reservoir design are outside the scope of this Manual. An excellent review of this topic is provided in McMahon and Mein (1978). Trends in hydrological data are discussed in the context of identifying inhomogeneities in streamflow data in Chapter 3 (section 3.3.3). However, this Manual does not describe the wide range of techniques available for trend detection which are fully described in WMO (2000). Although catchment modelling is presented in Chapter 9, primarily from the perspective of flow estimation at ungauged sites, the Manual does not present a detailed review, description or step-by-step procedures for model design, calibration, evaluation and validation. Beven (2001) reviews the principles and practice of hydrological modelling, and contains comprehensive references to their application. A similar approach is adopted in relation to low-flow forecasting. The key issues associated with different forecasting periods and forecasting methodologies are reviewed; however, a detailed step-by-step description of how forecasting models can be calibrated and operated in real time is also beyond the scope of this Manual. Although the Manual highlights the importance of ecological issues in lowflow design, it does not review the ecological processes and models used for setting instream flows. These are reviewed in detail in Tallaksen and van Lanen (2004). 1.5 . Guidelines for national practices. It is difficult to present standard guidelines for national practices for all aspects of low-flow estimation because of the diversity of hydrological environments and water resources problems, the extent of existing data, model availability and the time and skills needed to address a specific problem. Where possible, guidance for specific aspects of low-flow estimation has been provided. This is appropriate where a methodology is largely fully prescribed, for example, in calculating low-flow indices or fitting distributions to annual minima. However, many of the components for establishing national hydro-. 15. metric networks that underpin low-flow estimation, the development of statistical regional models, deterministic catchment models and comprehensive forecasting procedures cannot be summarized as a series of simple steps. Each specific approach will often involve choosing from a range of available options, which, in turn, requires the skills of the hydrological analyst and the experience gained from hydrological research and operational practice. This strategy can be applied usefully to developing national low-flow estimation procedures. There are clear advantages to having standard procedures formally approved by the national agencies responsible for water resources policy. However, this should not be extended to prescribing each and every aspect of low-flow estimation. Some useful guidance on the advantages and disadvantages of standardizing national flood design is given by Reed (1999). Much of this guidance is also relevant to low-flow estimation. References Beven, K.J., 2001: Rainfall-Runoff Modelling: The Primer. John Wiley & Sons, Chichester, United Kingdom. Gustard, A., A. Bullock and J.M. Dixon, 1992: Low Flow Estimation in the United Kingdom. Institute of Hydrology Report No. 108, Wallingford, United Kingdom. McMahon, T.A. and R.G. Mein, 1978: Reservoir Capacity and Yield. Developments in Water Science 9, Elsevier, Amsterdam, the Netherlands. Reed, D.W., 1999: Flood Estimation Handbook – Volume 1: Overview. Institute of Hydrology, Wallingford, United Kingdom. Tallaksen, L.M. and H.A.J. van Lanen (eds), 2004: Hydrological Drought – Processes and Estimation Methods for Streamflow and Groundwater. Developments in Water Science, 48, Elsevier Science B.V., Amsterdam. United Nations Educational, Scientific and Cultural Organization/World Meteorological Organization, 1992: International Glossary of Hydrology. Second edition, UNESCO, Paris. World Meteorological Organization, 1988, 2004, 2006: Technical Regulations (WMO-No. 49), Geneva. 2000: Detecting Trend and other Changes in Hydrological Data (Z.W. Kundzewicz and A. Robson, eds) (WMO/TD-No. 1013/WCDMP-45), Geneva. 2008: Guide to Hydrological Practices. Volume I, sixth edition (WMO-No. 168), Geneva (Volume II in preparation)..

(17) 16. ESTIMATING, PREDICTING AND FORECASTING LOW FLOWS. 2. Estimating, Predicting and Forecasting Low Flows 2.1 . Introduction. There are three primary scenarios when low-flow information is required. The first is when a water resources scheme is being developed. This normally requires a pre-feasibility study to determine whether the objectives of the proposal can be achieved. For schemes with a large capital investment, this is followed by much more detailed design work, including estimates of the frequency of low flows. The second is during the operational phase of a water resources scheme once it is constructed, and includes decisions on how to manage the scheme on a day-to-day basis. For example, it may be necessary to determine how much water can be diverted from a river for hydropower purposes without infringing legal abstraction conditions. These may depend on the time of the year or on the river discharge at a downstream point. In many countries, such operational constraints are not formalized and the operator must be aware of, and sympathetic to, the needs of many downstream water users. For example, water could be needed by households, or for agriculture, electricity production, navigation, industrial abstractions, tourism, the dilution of industrial or domestic effluent, maintaining an ecosystem for food production, or maintaining the natural biodiversity of a river. The third scenario is when it is necessary to make operational decisions today based on estimates of future river flows which look days, weeks or sometimes months ahead. These forecasts can increase the efficiency of water use and are of economic importance in terms of reducing the operational costs of water resources schemes. More general warnings of below average precipitation several months in advance are of value for long-term planning and for making contingency plans for severe droughts (Chapter 11). Historically, most applications of low-flow information have been in the design and operation of schemes for a specific water sector, such as public water supply, irrigation, energy, navigation and industry. It is now recognized that there is a need to provide long-term baseline monitoring and analysis of low flows to support integrated river basin management. This provides a. framework for environmental agencies to make decisions regarding the catchment-wide development of water resources and prevents ad hoc decisions from being made on a case-by-case basis. For example, by comparing current abstractions with estimates of low flows for all tributaries in a catchment, rivers that can no longer support further abstraction can be identified. Similarly, rivers that have potential for further abstraction without damaging downstream interests, including ecological demands, can be highlighted. Techniques for assessing water resources will be described in detail in the WMO Manual on Water Resources Assessment (in preparation). Long-term baseline information can also be used to give early warning of any natural or artificial change in the low-flow regime. 2.2 . Low-flow information. 2.2.1 Basin management Low-flow information is required for a wide range of applications that are often controlled by national and international water law and policy. For example, in Europe the Water Framework Directive was adopted by the European Parliament and the Council of the European Union in October 2000. This Directive established a strategic framework for the sustainable management of both surface water and groundwater resources. The Directive requires management by river basin authorities and not by administrative or political boundaries (see 12.3.2). A key component is to produce long-term river basin plans for integrated water resources management. This, in turn, will require estimates of low flows to mitigate the environmental impact of current abstractions and plan future water resources development. 2.2.2 River abstraction Traditionally, one of the most common uses of low-flow information has been the design and operation of public water-supply schemes. Information on the frequency of low river flows is required to assess the probability of the abstraction not meeting the anticipated demand. The abstraction could be direct to a water treatment plant or reservoir storage facility. In terms of hydrological.

(18) ESTIMATING, PREDICTING AND FORECASTING LOW FLOWS. analysis, the issues are similar for abstraction and irrigation, although there is normally a much higher seasonal and interannual (in temperate climates) variability in agricultural demand, which is dependent on crop type and the local climate. A frequent objective is to estimate the area that can be irrigated for a given crop type with a given risk of failure. Although the design of hydropower schemes is dependent on the complete range of flows, low flows can be critical in determining how much water must bypass a run-of-river hydroplant to maintain downstream river ecology and how much is available for power generation in the dry season. Thermal power stations are dependent on cooling water, and information on low flows when the availability of water for abstraction and the dilution of cooling water is at a minimum is essential for design purposes. For all these applications there may be a need to forecast flows in order to implement restrictions on water use to minimize the risk of very severe restrictions in the future. In some instances, licences to extract water in excess of the available supplies have been issued, and, thus, lowflow forecasts are an essential management tool. 2.2.3 Effluent dilution A common application of low-flow information is that of estimating the dilution of domestic or industrial discharge released into a river. A legal consent is frequently required to discharge a pollutant. Waterquality models based on the rate and quality of the discharge and the flow and quality of the receiving stream are used to determine the frequency distribution of downstream water quality. The flow-duration curve (cumulative distribution function) of receiving river flows is the most commonly used form of analysis (Chapter 6). This design application contrasts with the real-time application of assessing the impact of specific pollution incidents, for example, a discharge of oil after an industrial accident. Flow rates are required to estimate the rate of dispersion and time of travel so as to warn downstream public water-supply abstractors that abstraction must be ceased. 2.2.4 Navigation River systems provide an important transport facility for both industrial and leisure navigation. Navigation is interrupted during low-flow periods because the reduced water levels cannot accommodate vessels and the water available is insufficient for locks to be operational. The critical hydrological variable is water depth, and, in the absence of field observations, the depth profile must be estimated using the time series of river flows and hydraulic models. These enable the frequency. 17. of interrupted navigation for different vessel sizes to be estimated and proposals can be made for improved channel design, while protecting the natural ecosystem. Navigation is driven by long-term investment and the related infrastructure cannot be easily relocated, redesigned or reconstructed. It is therefore important to forecast low flows so that shipping agencies are warned of navigation restrictions and have an opportunity to provide alternative means of transport in extreme conditions. 2.2.5 Ecosystem protection and amenities Ecosystems are most vulnerable during low-flow periods because of a reduction in the availability of habitats, water temperature extremes, a reduction in dissolved oxygen, a deterioration in water quality (owing to reduced effluent dilution) and habitat fragmentation (caused by natural or artificial barriers to fish movement). Related techniques include simple methods based on low-flow indices often known as “standard setting” and include the mean annual minima of a given duration (Chapter 5) or a percentile from the flowduration curve (Chapter 6). The estimated discharge is then used to set a minimum flow in a river so that when the discharge falls below this level, abstractions should cease (or be reduced). More advanced methods include habitat modelling where the impact of specific flow regimes on different species and life stages are assessed (Tallaksen and van Lanen, 2004). In addition to supporting complex ecosystems, rivers are natural assets for sport and recreational activities (for example, canoeing, fishing, ornithology and walking). Ensuring adequate water depth or velocity even when flow rates are very low can artificially enhance the natural appeal of rivers. 2.3 . Design issues. 2.3.1 Estimation, risk and forecasting This Manual employs the general term “low-flow estimation”. This may be applied to a specific measured discharge on a given day, a low-flow statistic calculated from data, a statistic estimated from a model or a forecast of future low flows. The term “estimation” implies that there is a calculation error and that the predicted value will differ from the observed one. The WMO Guide to Hydrological Practices (WMONo. 168) considers estimation errors in detail. Low flows can be analysed in a number of different ways. Some are single values such as a recession constant, the proportion of base flow or the mean of a series..

(19) 18. ESTIMATING, PREDICTING AND FORECASTING LOW FLOWS. These are called low-flow indices. More complex methods estimate low-flow probability. For example, the cumulative frequency distribution (flow-duration curve) of daily flows describes the relationship between discharge and the percentage of time that a specific discharge is exceeded (Chapter 6). Extreme value techniques are used in Chapter 7 to estimate the nonexceedance probability of annual minima. The essential difference in the two techniques is that the flow-duration curve considers all days in a time series and hence the percentage of time over the entire observational period that a flow is exceeded. In contrast, the extreme value techniques applied to annual minima data estimate the non-exceedance probability in years, or the average interval in years (return period) when the annual minima are below a given value. It is therefore often helpful to specify the design life of a particular water resources scheme. The risk r of experiencing one or more annual minima lower than the T-year minimum discharge during a design life of M years is: r = 1–(1–1/T) M . (2.1). forecasts have a typical forecast lead time of less than seven days; medium-term forecasts cover up to six months, and long-term forecasts over six months. 2.3.2 Annual, seasonal and different durations Low-flow information has traditionally been based on estimating statistics from all the available data. However, for many applications, it may be more appropriate to consider data according to months, groups of months or specific seasons. For example, in designing an irrigation scheme, the analysis should focus on the period of the year when the abstraction for irrigation will take place, and data for the rest of the year may be redundant. Similarly, ecological modelling may establish that the flow for a particular species and life stage are important, and, once again, the analysis should focus on a critical seasonal period. Annual minimum and flow-duration analysis can both be carried out for specific months or groups of months. It may be appropriate to consider annual minima of different durations, for example 7-day, 10-day, 30-day and 90-day durations, to meet the requirements of a specific design problem.. The risk probability r of experiencing the 100 year annual minima during a design life of 100 years can be estimated by setting T = 100 and M = 100 in the above equation. This gives a probability of 0.63 and provides a useful approach for demonstrating that the probability of “failure” during the life of any scheme is significant and that contingency plans for extreme events should be incorporated into design aspects. Although the design of water resources schemes is outside the scope of this Manual, it should be noted that a key concept of any such scheme is the “level of service”. This is an estimate of how frequently and for how long the planned demands of a water resources scheme user will not be met. For example, an irrigation scheme may provide only 50 per cent of the full design abstraction for three months, once every 20 years. Low-flow estimation is a key component in estimating the level of service.. 2.3.3 Scales of estimation A key issue in operational hydrology is the management unit, which could be the river network or infrastructure (for example, reservoirs, aqueducts, pumps, water treatment plants, distribution networks, sewerage treatment plants). The primary management unit for environmental protection is the river basin, and the importance of upstream hydrology and upstream developments on downstream issues has highlighted the need for integrated river basin management. For hydropower generation, the management unit is often the reservoir and turbine facilities, with some consideration given to release regimes to protect downstream interests. Water-supply management units are often very complex and may involve the combined use of surface water and groundwater systems or the reuse of sewage effluent. Such systems require complex hydrological, water resources and water use modelling.. These methods do not attempt to estimate when a specific discharge or low-flow statistic will occur. In contrast, the forecasting techniques described in Chapter 11 estimate both the magnitude of low flow and the time of the event days or months ahead. As forecast lead time increases, forecast accuracy decreases, and, for very long lead times, forecasts may be no more accurate than those made using the long-term statistical mean. Forecasting methods are categorized according to the time interval for which the forecast is made. Short-term. Water resources problems occur over a wide range of space scales. These range from detailed estimates at individual reaches of the order of 100 m within 10 km2 catchments to the estimation of low-flow frequency at the river basin scale covering areas in excess of 1 million km2. In developing countries, for large river basins in excess of 1 000 km2, rivers are normally gauged with long (greater than 50 years) time series of daily flow data. This enables design problems to be most commonly based on an analysis of gauged data..

(20) ESTIMATING, PREDICTING AND FORECASTING LOW FLOWS. However, these larger catchments often present the interesting challenge of separating the myriad of ar-tificial influences from the natural flow regime. In developing countries, however, because of a lack of good-quality and continuous observational data, many key river basins are ungauged and low-flow analysis must therefore be based on model estimates. Even in countries with dense river-gauging networks, there are a large number of ungauged river reaches. This has led to the development of a wide range of regional models for estimating low flows at ungauged sites. Lastly, to compare the resources of countries on the continental scale, reliable estimates of resources availability determined using a consistent methodology are needed. Simple hydrological models are appropriate in these situations because the key requirement is to identify the spatial variability of the resources, and a simple low-flow index is often sufficient. Low-flow investigations may take place on the scale of the river reach, the catchment, national and international basins, the region (group of countries) or globally. The scale of the study will be a primary influence on the approach adopted and the information required. 2.3.4 The low-flow cube Hydrologists can adopt a variety of procedures for hydrological design and water management, the selection of which is determined by the nature of the output required – the design requirement. The choice is determined by the risk associated with the design decision. For example, the investment risk linked to the construction of a large impounding reservoir would necessitate the establishment of gauging stations and the analysis of observed river flows. These data provide the basis for hydrological design, typically the storage/yield characteristics and spillway capacity. In contrast, an application for a small-scale abstraction licence would frequently be at an ungauged location, not warranting gauging station construction, and the design would often be based on a flow-regime statistic, such as the dry season flow of a given reliability, without necessarily requiring timeseries analysis. The complexity of different design scenarios (UNESCO, 1997) can be simplified by the conceptuali-zation of a “Design Scenario Cube” (Figure 2.1), defining three dimensions to the design requirement, as follows:. 1. The location of the design problem, for example, at a site where recorded hydrological data is not available. Alternatively, there may be a nearby upstream or downstream gauging station. This distinguishes between the gauged and ungauged situation.. 19. 2. The operational requirements of the hydrological design. For example, the water sector and magnitude of the capital investment will determine whether simple statistics are required or a long and continuous hydrological time series. This is the data characteristics dimension that distinguishes between the requirement for timeseries or low-flow statistics.. 3. The catchment may be relatively natural or heavily modified by water resources development, in which case it may be necessary to naturalize the flow record (Chapter 3). The catchment water-use dimension distinguishes between the requirements for natural or artificially influenced flows.. The three-dimensional cube presents eight distinct combinations of different design scenarios. Each one is summarized below:. 1. Natural gauged time series: A time series of (daily) flows at a gauged site which represent the riverflow regime from a natural catchment. An application of this design requirement may be in setting environmental river-flow regimes at a location in the vicinity of a gauging station.. 2. Artificial gauged time series: A time series of daily flows at the gauged site which represent the river-flow regime from an unnatural, artificially influenced catchment. Different definitions of “artificial” can be adopted including historic, current or future scenario-based water use. An application of this design requirement is to evaluate the extent to which different upstream abstractors have reduced downstream flows over a historical period.. 3. Natural gauged statistics: A flow regime statistic, such as the 95 percentile flow, which represents the river-flow regime from a natural catchment at a gauged site. This may be needed to determine whether to approve a proposed abstraction.. 4. Artificial gauged statistics: A flow regime statistic that represents the river-flow regime from an artificially influenced catchment at a gauged site. This design may be required to determine the extent to which upstream abstractors have diminished the 95 percentile low-flow discharge at a gauging station..

(21) 20. ESTIMATING, PREDICTING AND FORECASTING LOW FLOWS. 2.4 Statistics Time series Gauged. Ungauged Natural. Artificial. Figure 2.1. Design Scenario Cube: A conceptualization of the variety of approaches to estimate low flows. 5. Natural ungauged time series: A time series of river flows that represents the natural regime at an ungauged site. This would be needed to design a complex abstraction scheme at an ungauged site and would require the development of a catchment simulation model (Chapter 9).. 6. Artificial ungauged time series: A time series of river flows which represents an artificially influenced regime at an ungauged site. A typical application of this design requirement may be in the design of a joint use (surface water and groundwater) scheme in a catchment that has groundwater pumping but no discharge measurements.. 7. Natural ungauged statistics: A flow regime statistic that represents the natural regime at an ungauged site. This may be required for preliminary regional water resources assessment.. 8. Artificial ungauged statistics: A flow regime statistic that represents the artificially influenced regime at an ungauged site. A typical application of this requirement may be in the preliminary design of a small-scale hydropower scheme in an artificially influenced catchment.. The added complexity of this scheme includes evaluating the impact on downstream river flows of artificial influences based on historic sequences, current water use or future abstraction scenarios. In most cases, this will require the naturalization of a continuous flow record or low-flow statistics (Chapter 10).. Previous studies. Before embarking on a low-flow investigation, it is essential to review previous work carried out in the region. Procedures for collating this information are described in detail in WMO (2008). Such procedures should include a review of all hydrological data. Also, if methods for estimation at an ungauged site are being considered, then thematic data (Demuth, 1993) describing the climate, soil, geology and geomorphology should also be collated. For example, although a regional low-flow study may not have been carried out in a region, a flood study might have compiled a useful set of catchment descriptors (Chapter 3). The main sources of data that should be reviewed include: hydrometeorological data, physiographic data, anthropogenic data, hardcopy and digital maps and previous hydrological studies published in reports or journals. This information may be available from metadata catalogues, project archives, libraries, personal communications and the Internet. 2.5 . Dissemination of results. A key component in any low-flow investigation is to determine at the outset how the results will be disseminated to the user community (WMO, 2008). Users may range from hydrological specialists familiar with lowflow analysis methods, to professionals in the disciplines of engineering, ecology or law, to the general public with an appreciation of the natural environment. Traditionally, results have been provided in report and map format; but, this is now being supplemented or replaced by software for calculating regional low-flow estimates and the digital display of results or online bureau services. This is possible because of the increasing operational use of digital hydrological information including elevation models and stream networks (Chapter 3). There is still a need to ensure that low-flow methods and results are fully understood, and it is essential for online dissemination to be supported by comprehensive training in the methods and application of low-flow information. 2.6 . Key principles in low-flow design. Several general principles in hydrological design (Reed, 1999) are contained in this Manual and should be considered by hydrologists when carrying out low-flow design for a specific scheme or when developing national estimation procedures. The common objective of all design procedures is to reduce the uncertainty of lowflow estimation. Some general principles are as follows:.

(22) ESTIMATING, PREDICTING AND FORECASTING LOW FLOWS. (a) L ow-flow estimation should, wherever possible, be based on recorded data;. (b) In the absence of recorded data at the site, data transfer upstream or downstream or from a nearby catchment will lead to reduced estimation error;. (c) The estimation of low-flow statistics at ungauged sites from catchment descriptors may be enhanced by some kind of formal or informal data transfer;. (d) The ability to select the most appropriate lowflow analysis and estimation method is a matter of experience. The decision will always be influenced by the nature of the design problem, the catchment, the availability of data and the experience of the practitioner;. (e) The enquiring analyst can usually find more information, which, if used, will lead to reduction in estimation error; (f) Given the growing automation of modern analysis and estimation software, professionals are being increasingly obliged to take more care in hydrological design work and to always seek advice from a professionally qualified expert; (g) Positive and negative feedback from a user community, together with advances in methodologies and data availability, enables minor. 21. revisions to national design procedures and software. It is important for any errors or shortcomings of low-flow estimation techniques to be rectified immediately. However, continual minor adjustments to design methods are not welcomed by the user community, which generally prefers major updates to be less frequent.. References Demuth, S., 1993: Untersuchungen zum Niedrigwasser in West-Europa. Freiburger Schriften zur Hydrologie, Bd. 1. Reed, D.W., 1999: Flood Estimation Handbook. Volume 1, Institute of Hydrology, Wallingford, United Kingdom. Tallaksen, L.M. and H.A.J. van Lanen (eds), 2004: Hydrological Drought – Processes and Estimation Methods for Streamflow and Groundwater. Developments in Water Science, 48, Elsevier Science B.V., Amsterdam. United Nations Educational, Scientific and Cultural Organization (UNESCO), 1997: Southern Africa – FRIEND. Technical Documents in Hydrology No. 15, UNESCO, Paris. World Meteorological Organization, 2008: Guide to Hydrological Practices. Volume I, sixth edition (WMO-No. 168), Geneva (Volume II in preparation)..

(23) 22. HYDROLOGICAL DATA. 3. Hydrological Data. 3.1 . Introduction. Hydrological and related data encompasses all data commonly used by hydrologists (WMO, 1958, 2006). Meaningful hydrological analysis will always rely on good-quality data that are representative of conditions in the drainage basin, study area or region concerned over the period of interest. The aim of this chapter is to introduce the reader to the data types most commonly used in low-flow hydrology and provide guidance on preparing such data for analysis. This chapter emphasizes the preparation of river-flow (or streamflow) data, because of their specific importance in low-flow studies. However, the chapter deliberately excludes details on hydrometry, hydrometric data processing and management, network design, and the determination and management of hydrological data: several WMO technical publications and reports (for example, WMO, 1980, 2008a) and other publications (for example, Herschy, 1999) already cover these topics adequately. Where necessary, reference is made to relevant publications. The WMO Manual on Water Resources Assessment (in preparation) will provide specific details on the collection and processing of biophysical data (topography, soil, geology and vegetation), socio-economic data (land use and demography) and climate data (precipitation and evaporation). The chapter provides a summary of some of these variables. 3.2 . Data for low-flow analysis. The relevant data for low-flow analysis may be grouped as follows: hydrometeorological, physiographical or anthropogenic. Hydrometeorological data describe elements of the hydrological (water) cycle which continually vary over time. Physiographical data, on the other hand, represent natural terrestrial features that do not change, or vary insignificantly, over the relatively short timescales considered by most hydrological analyses. Anthropogenic data help explain the influences of human activity on both natural and artificial (man-made) systems.. Most hydrological data may be further subdivided into point measurements and spatial data. A point measurement describes the measurement of a certain entity at a specific location. A sequence of measurements, samples, observations or readings of a time-varying entity at a single location results in a time series of that entity at that point. Time-series data can be manipulated to produce further time series of aggregated data and statistics. For example, calculating the average of 96 15-min streamflow measurements during one day gives a single daily mean flow value. Repeating this for several days’ data produces a time series of daily mean flow values. Further aggregation of the daily data could produce weekly, 10-day, monthly, or annual time series. A wide range of statistics (for example, means, maxima, minima, variance) can be determined for different periods covered by the time series to provide a useful summary of the data and form the basis of analysis. To remove short-term fluctuations and study the general behaviour of data, it is sometimes useful to generate a moving average through a time series. In low-flow hydrology, 7-, 30- or 120-day moving averages, derived from daily data, are frequently used, for example, to calculate 7-day annual minimum flow, for a single year, or the mean of the 7-day annual minima (MAM(7)), for several years. Spatial data can simply be defined as any data that can be represented on a map. Spatial data are often derived from the interpolation of many point measurements. The contours on a topographical map, for instance, are usually derived from interpolating a series of spot elevation readings. When digitized (namely, stored in digital form on a computer), line, arrow, contour, point and polygon features of a map are referred to as vector data. Computerized grid-, cell- or lattice-based information, such as a digital elevation model, is generically called raster data. For convenience and ease of analysis, digital spatial data are normally stored on computer as data coverages in geographic information systems (GIS). Some examples of the different types of data that a low-flow hydrologist may use are given in the following subsections..

(24) HYDROLOGICAL DATA. 3.2.1 Hydrological data Hydrological data include (WMO. 2006) hydrometric, groundwater and climatological data for hydrological purposes. They provide information on the spatial and temporal distribution of water, as it occurs in its various states, in the hydrological cycle. Here, only brief descriptions are given of the types of hydrological data most commonly used in low-flow studies. Hydrometric data Hydrometric data include streamflow and river- and lake-level (stage) data. Of all the different types of hydrological data, streamflow (synonymously riverflow or discharge) data are arguably the most useful in low-flow hydrology because they represent the combined response of all physical processes operating in the upstream catchment (Herschy, 1995). River flow is the rate at which water flows through a given river cross-section and is usually expressed in units of m³/s, l/s or ft³/s. Several different methods can be used to measure river flow, with the choice of method depending on the conditions at a particular site (WMO, 2008a). The methods generally involve the measurement of water level, or stage, at a gauging station and the subsequent application of a stage-discharge relationship to derive a flow estimate. Most countries have long-established hydrometric networks comprising many gauging stations, from which data are used for a variety of purposes ranging from water resources management and planning to flood control, environmental monitoring and impact assessment. However, in all countries, gauging stations contribute unevenly to the understanding of low flows, with the most valuable stations being those with the best hydrometric performance, the least human disturbance and a long time series (Rees and others, 2004). The reader is referred to the WMO Manual on Stream Gauging (WMO-No. 519), the WMO Guide to Hydrological Practices (WMO-No. 168), technical guidance manuals of the International Organization for Standardization (ISO), such as Liquid Flow Measurement in Open Channels (ISO, 1981, 1982) and textbooks (for example, Herschy, 1995, 1999) for further information on hydrometric practices and gauging station network design. The WMO INFOHYDRO Manual (WMO-No. 683) should be consulted for information on the status of networks, and data availability, in different countries.. 23. Climatological data for hydrological purposes This data include measurements of precipitation, evapotranspiration, air temperature, radiation, wind, humidity and barometric pressure, and synoptic data, which describe weather features, such as high- and low-pressure areas and weather fronts, and atmospheric circulation patterns and indices (for example, El Niño/Southern Oscillation and the North Atlantic Oscillation Index). Information on methods for observing data can be found in the WMO Guide to Meteorological Instruments and Methods of Observation (WMO-No. 8). Precipitation falls either in liquid form, as rain, drizzle or dew, or in solid form, as snow, sleet, hail, hoar frost or rime. Knowledge of precipitation distribution in time and space is valuable in any low-flows analysis. The total amount of precipitation reaching the ground during a given period is expressed as the depth that would cover a horizontal projection of the Earth’s surface (WMO, 2008a). Precipitation is most commonly sampled by individual precipitation gauges (raingauges, if only liquid precipitation is measured), at varying intervals, depending on the equipment and measurement method. Precipitation gauges are sensitive to exposure, particularly wind, and the amount measured may be less than the actual precipitation by up to 30 per cent or more (WMO, 2008b). The systematic error of precipitation varies according to the type of precipitation (measurement errors for solid precipitation are often an order of magnitude greater than those normally associated with liquid precipitation), the location of the gauge (exposure or shielding at the site) and the instrumentation used (for example, gauge type, orifice height or diameter). For many studies it may be necessary to adjust observed data to allow for systematic errors. Care must always be taken when adjusting data, and changes should always be adequately documented. The WMO Operational Hydrology Report No. 21, Methods of Correction for Systematic Error in Point Precipitation Measurement for Operational Use (WMO-No. 589), describes methods to adjust point precipitation measurements. Observed precipitation is representative only of precipitation falling over a limited area in the immediate vicinity of the gauge, and, to understand the spatial distribution of precipitation over a wider area (for example, a catchment or drainage basin), data should be obtained from a network of gauges. A variety of interpolation methods can be applied to raingauge measurements to produce isohyetal maps, showing lines (contours) of equal rainfall (isohyets), or grids of rainfall..

No results found